@inproceedings{ahmad-etal-2026-open,
title = "{OPEN}-{SWE}-{TRACES}: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents",
author = "Ahmad, Wasi Uddin and
Ludwig, Nikolai and
Majumdar, Somshubra and
Ginsburg, Boris",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.131/",
pages = "2668--2679",
ISBN = "979-8-89176-462-0",
abstract = "The path toward autonomous software engineering is currently bottlenecked by a severe deficit of diverse, large-scale trajectory data. We address this by introducing , an expansive dataset of 207,489 agentic trajectories spanning nine programming languages (Python, Go, TS, JS, Rust, Java, PHP, C, C++). Sourced from 20,000 real-world PRs via OpenHands and SWE-agent harnesses, the dataset utilizes a hybrid-reasoning synthesis: Minimax-M2.5 generates trajectories with explicit ``thinking'' processes, while Qwen3.5-122B provides high-quality behavioral traces. Filtered for permissive licenses (MIT, Apache, BSD) from SWE-rebench-V2, this data facilitates the training of models capable of long-horizon reasoning. We validate the dataset by fine-tuning the Qwen3-30B-A3B series (Thinking, Coder, and Instruct). The best performing model achieves a resolve rates of 60{\%} on SWE-bench Verified, 46.2{\%} on SWE-bench Multilingual, and 36.8{\%} on SWE-bench Pro. These results establish as a premier resource for distilling human-level software engineering capabilities into efficient, open-source agentic LLMs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ahmad-etal-2026-open">
<titleInfo>
<title>OPEN-SWE-TRACES: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wasi</namePart>
<namePart type="given">Uddin</namePart>
<namePart type="family">Ahmad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolai</namePart>
<namePart type="family">Ludwig</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Somshubra</namePart>
<namePart type="family">Majumdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Boris</namePart>
<namePart type="family">Ginsburg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>The path toward autonomous software engineering is currently bottlenecked by a severe deficit of diverse, large-scale trajectory data. We address this by introducing , an expansive dataset of 207,489 agentic trajectories spanning nine programming languages (Python, Go, TS, JS, Rust, Java, PHP, C, C++). Sourced from 20,000 real-world PRs via OpenHands and SWE-agent harnesses, the dataset utilizes a hybrid-reasoning synthesis: Minimax-M2.5 generates trajectories with explicit “thinking” processes, while Qwen3.5-122B provides high-quality behavioral traces. Filtered for permissive licenses (MIT, Apache, BSD) from SWE-rebench-V2, this data facilitates the training of models capable of long-horizon reasoning. We validate the dataset by fine-tuning the Qwen3-30B-A3B series (Thinking, Coder, and Instruct). The best performing model achieves a resolve rates of 60% on SWE-bench Verified, 46.2% on SWE-bench Multilingual, and 36.8% on SWE-bench Pro. These results establish as a premier resource for distilling human-level software engineering capabilities into efficient, open-source agentic LLMs.</abstract>
<identifier type="citekey">ahmad-etal-2026-open</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.131/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>2668</start>
<end>2679</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OPEN-SWE-TRACES: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents
%A Ahmad, Wasi Uddin
%A Ludwig, Nikolai
%A Majumdar, Somshubra
%A Ginsburg, Boris
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F ahmad-etal-2026-open
%X The path toward autonomous software engineering is currently bottlenecked by a severe deficit of diverse, large-scale trajectory data. We address this by introducing , an expansive dataset of 207,489 agentic trajectories spanning nine programming languages (Python, Go, TS, JS, Rust, Java, PHP, C, C++). Sourced from 20,000 real-world PRs via OpenHands and SWE-agent harnesses, the dataset utilizes a hybrid-reasoning synthesis: Minimax-M2.5 generates trajectories with explicit “thinking” processes, while Qwen3.5-122B provides high-quality behavioral traces. Filtered for permissive licenses (MIT, Apache, BSD) from SWE-rebench-V2, this data facilitates the training of models capable of long-horizon reasoning. We validate the dataset by fine-tuning the Qwen3-30B-A3B series (Thinking, Coder, and Instruct). The best performing model achieves a resolve rates of 60% on SWE-bench Verified, 46.2% on SWE-bench Multilingual, and 36.8% on SWE-bench Pro. These results establish as a premier resource for distilling human-level software engineering capabilities into efficient, open-source agentic LLMs.
%U https://aclanthology.org/2026.findings-emnlp.131/
%P 2668-2679
Markdown (Informal)
[OPEN-SWE-TRACES: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents](https://aclanthology.org/2026.findings-emnlp.131/) (Ahmad et al., Findings 2026)
ACL